Diabetologia (1992) 35:347-356
Diabetologia 9 Springer-Verlag 1992
Metabolism of HDL apolipoprotein A-I and A-II in Type 1 (insulin-dependent) diabetes mellitus M.-R. T a s k i n e n 1, J. Kahri 1, V. K o i v i s t o 2, J. Shepherd 3 and C.J. Packard 3
1Third and 2Second Departments of Medicine, University of Helsinki, Finland and the 3Institute of Biochemistry, Royal Infirmary, Glasgow, Scotland, UK
Summary. Concentrations of HDL cholesterol and apolipoprotein A-I are commonly increased in Type 1 (insulindependent) diabetes mellitus but the mechanisms whereby diabetes influences H D L metabolism have not been studied. We investigated the metabolism of HDL apoproteins A-I and II in normolipidaemic Type 1 diabetic men (n = 17, HbA~ 6.4-11.9 % ) without microalbuminuria but with a wide range of H D L cholesterol (0.85-2.10 mmol/1) and in nondiabetic men (n = 18) matched for body mass index and the range of H D L cholesterol. Input rates and fractional catabolic rates for apolipoproteins A-I and II were determined following injection of ~2q-apolipoprotein A-I and ~31I-apolipoprotein A-II tracers. Additional multicompartmental analysis was performed using a model to describe the kinetics of HDL particles containing only apolipoprotein A-I (Lp A-I) and apolipoprotein A-I and apolipoprotein A-II (Lp A-I/ A-II). No gross differences from normal subjects were observed in the mean levels of lipids, lipoproteins, apoproteins and the lipolytic enzymes in the diabetic men as a result of the selection process. Furthermore, the relationship between
H D L cholesterol, apolipoprotein A-I, apolipoprotein A-II, kinetic analyses, VLDL triglyceride, lipolytic enzymes.
A reduced H D L cholesterol level in plasma is a recognised risk factor for coronary heart disease [1]. The concentration and composition of this lipoprotein are regulated by the combined actions of a number of factors including lipoprotein lipase (LPL), hepatic lipase (HL) and cholesteryl-ester transfer protein (CETP) [2, 3]. In addition plasma H D L levels are known to be influenced by changes in the synthesis and catabolism of the major H D L proteins, apolipoproteins (apo) A-I and A-II. H D L turnover studies have shown that the fractional catabolic rate (FCR) of apo A-I correlates inversely with the H D L levels in the circulation [4-7], i. e. a slow apo A-I F C R is associated with a high concentration of apo A-I and H D L . In contrast, fast clearance of apo A-I leads to low H D L levels regardless of the presence or absence of hypertriglyceridaemia [7-9]. Since coronary heart disease is the principal complication and leading cause of death in diabetes, H D L has been
a focus of great interest in both the insulin-dependent and non-insulin-dependent forms of the condition. In general H D L cholesterol is lower in Type 2 (non-insulin-dependent) patients compared to non-diabetic populations matched for sex, age and body mass index [10]. Conversely, most studies have reported that in Type i diabetic patients with fair to good glycaemic control, the concentration of H D L cholesterol is either increased or normal [10]. Lowering of H D L cholesterol has been documented in two studies [11, 12]. Thus, the response of H D L in diabetes seems to be variable. The observed elevation of H D L cholesterol in Type i diabetes is mainly due to high levels of HDL2 but a rise of HDL3 has also been reported [13-16]. Not unexpectedly, therefore, the concentration of apo A-I in the plasma is generally elevated whereas that of apo A-II seems to be within the normal range [17-19]. Hyper-apo A-I is considered to be anti-atherogenic [3]. The apo A-I/A-II ratio rises consistently with a preferen-
apolipoprotein A kinetics and plasma HDL cholesterol levels appeared to be preserved in the diabetic group. However, some normal interrelationships were disrupted in the diabetic men. Firstly, the rate of apolipoprotein A-II synthesis was 22% lower than in control subjects (p < 0.05). Modelling indicated that this was due to decreased input of Lp A-I/A-II particles whereas the input of Lp A-I particles was similar in the two groups. Secondly, there was no correlation between VLDL triglyceride and H D L cholesterol or VLDL triglyceride and the fractional catabolic rate of apolipoproteins A-I and A-II in diabetic men in contrast to that seen in control subjects. We conclude that there is a disruption in the normal association between VLDL and HDL metabolism in Type i diabetic men and postulate that the observed differences may be due to the therapeutic use of exogenous insulin. Key words: Type 1 (insulin-dependent) diabetes mellitus,
348 tial e l e v a t i o n of HDL2 which is the a n t i - a t h e r o g e n i c subclass of H D L cholesterol [3, 10]. C o n s e q u e n t l y there is an u n s o l v e d p a r a d o x in Type 1 diabetes which is in g e n e r a l associated with a n increased risk of c o r o n a r y h e a r t disease. O v e r a l l the m e c h a n i s m s by which diabetes m o d i f y H D L m e t a b o l i s m are n o t fully u n d e r s t o o d . We have s h o w n that in Type i diabetes, as in the n o n - d i a b e t i c p o p u l a t i o n , lipolytic e n z y m e s ( L P L a n d H L ) regulate p l a s m a HDL2 [14]. T h e o b s e r v e d e l e v a t i o n of HDL2 cholesterol c a n be at least partly e x p l a i n e d b y a rise of L P L activity due to p e r i p h e r a l h y p e r i n s u l i n i s m in c o n v e n tionally t r e a t e d Type 1 diabetes [20]. However, there are n o reports o n the m e t a b o l i s m of apo A - I a n d A - I I in Type 1 diabetic patients. I n the p r e s e n t study we have e x a m i n e d in vivo the t u r n over of apo A - I a n d A - I I in n o r m o l i p i d a e m i c Type 1 diabetic subjects with fair to good glycaemic control a n d a wide r a n g e of H D L cholesterol. I n particular, we sought to d e t e r m i n e w h e t h e r the diabetic c o n d i t i o n per se inf l u e n c e d H D L m e t a b o l i s m in Type 1 diabetes. Consequently, the p a t i e n t s were c o m p a r e d to n o n - d i a b e t i c control subjects with a similar wide r a n g e of H D L levels. I n addition, we m e a s u r e d the activities of lipolytic e n z y m e s ( L P L a n d H L ) in p o s t - h e p a r i n i z e d plasma.
Subjects and methods Seventeen Type i diabetic men selected to provide a wide range of HDL cholesterol (0.85-2.10 retool/l) but with normal serum triglyceride concentrations ( < 2.5 mmol/1) participated in the study. The average age of the patients was 36_+2 years (mean + SEM, range 23-59 years) and the mean body mass index ranged from 18.1 to 27.4 kg/m 2(mean _+SEM, 23.5 + 0.6 kg/m2). The duration of diabetes averaged 10.6 + 1.8 years (range 2-25 years). Sixteen patients received insulin by injection, two (n = 5) to three (n = 11) times daily and in one patient it was administered subcutaneously via a pump. The average dose of daily insulinwas 41 + 3 IU which remained constant throughout the study. The concentrations of HbAl and fructosamine ranged from 6.4 to 11.9% and 2.2 to 3.9 gmol/1 respectively. All the patients had C-peptide values of less than 0.33 nmol/1. Overnight albumin excretion rates (3-18 gg/min) were normal as was the mean serum creatinine at 94 + 11 (mean + SEM) gmol/1. Two patients had background retinopathy. None was taking lipid-lowering drugs or other drugs known to influence lipid metabolism. The patients were instructed to follow a weight-maintaining, sucrose-free diet containing 45 % of its calories as carbohydrate, 35 % as fat and 20% as protein. The patients were recruited from the Diabetes Clinic of the Helsinki University Central Hospital. Control men (n = 18) were recruited from volunteers screened for the Helsinki Heart Study between 1982 and 1983 but excluded from the study because their non-HDL cholesterol was < 5.2 mmol/1. A subgroup (n = 92) of these men living in the Helsinki area were screened again in 1988. The control group was selected to have similar ranges of HDL cholesterol and serum triglyceride as the Type 1 diabetic patients. None had a history of cardiovascular disease. Endocrine and other disorders which could influence lipoprotein metabolism were excluded by medical history, clinical examination and liver, kidney and thyroid function tests. None of the subjects was taking any medication. The absence of diabetes was confirmed by measurement of fasting blood glucose and serum fructosamine. All clinical and laboratory work was cornpleted in Helsinki whereas the kinetic analyses of apo A-I and A-II were performed in Glasgow. The study protocol was approved by the Ethical Committee of Helsinki University Hospital and informed consent was obtained from each subject.
M.-R. Taskinen et al.: Metabolism of apo A-I and A-II in diabetes
Purification and labelling of apolipoprotein A-I and A-H Heterologous apo A-I and A-II were used to guarantee the same batch and structure of apoprotein. Apo A-I and A-II were prepared in Glasgow from the plasma o f healthy donors, who were negative on screening for hepatitis B and human immunodeficiencyvirus. HDL was separated by ultracentrifugationand apo A-I and A-II prepared after delipidation and purification using high performance liquid chromatography as described previously [21]. The purity of apoproteins was established by a battery of immunoassays and by amino acid analysis [22]. The isolated apoproteins were dialysed against 0.1mot/I NH4HCO3 (pH 8.6), lyophilized and stored at -70~ Batches of lyophilized apo A-I and A-II were mailed to Helsinki in dry ice and stored at -70~ Apo A-I (0.5mg) and apo A-II (0.5 mg) were trace labelled with i mCi of 1~3Iand I mCi 131Irespectively, by the iodine monochloride method [21, 23]. Radioiodinated apoproteins were freed from unbound radioiodide by column chromatography (Sephadex G-25M, Pharmacia, Uppsala, Sweden) and used immediately.
Kinetic studies The subjects were examined as out-patients to avoid disruption of their daily life. No alcohol intake was allowed during the study and the subjects were instructed to maintain their normal physical activity. For 3 days before and throughout the study the subjects received potassium iodide (60 rag, three times per day) to minimise thyroid absorption of radioactive iodide. Fasting blood was drawn into sterile tubes containing 300 B1 of 0.40 mol/l disodium-EDTA solution in 80ml samples. Plasma (25 ml) was adjusted to a density of 1.063 g/ml by adding solid KBr and centrifuged in a Beckman Ti 60 rotor (24 h, 40000 rev/min, 4 ~ The top fraction containingVLDL, intermediate density Iipoprotein (IDL) and LDL was removed and the density of the bottom adjusted to 1.210 g/ml by addition of solid KBr. HDL was isolated by a further centrifugation for 24 h as above. The HDL fraction was harvested by aspiration of 1-2 ml samples and stored at 4 ~ for up to 1-2 h. 125I-labelledapo A-I and 131I-labelledapo A-II were incubated with 0.5-1 ml of the HDL of each subject for 30 rain at room temperature. The density of the incubation mixtures was adjusted to 1.21 g/mI by using solid KBr, overlayered with 3.5 ml of KBr solution (density = 1.210 g/ml) and the HDL re-isolated by ultracentrifugation for 21 h at 40000 rev/min at 4~ in a Beckman Ti 50.3 rotor. ~25Iapo A-I HDL and 131Iapo A-II HDL were harvested by aspiration and dialysed extensively against sterile 0.15 mol/1 NaC1, 0.01% disodium-EDTA. Thereafter the preparations were sterilised by membrane filtration (0.22 gmol Millipore filters, Millipore, Bedford, Mass., USA). On the fifth morning after the collection of the initial blood samples the subjects received 25 gCi each of labelled apo A-I HDL and apo A-II HDL by i.v. bolus injection. The first blood sample was taken after 10 rain and then daily after a 10-h overnight fast until day 13.24-h urine specimens were collected over 13 days and the urine volume measured. The concentration of creatinine in the urine was used as an index of completeness of collection. The radioactivity was measured in 2-ml aliquots of plasma and urine, and was counted at the end of the study.
Lipid, lipoprotein and apoprotein determinations On days 1,6, 10 and 13 of the turnover period blood was drawn after a 10-h overnight fast for measurement of serum lipids and lipoproteins. Lipoprotein fractions were isolated by sequential ultracentrifugation [24] in a Beckman L7-70 ultracentrifuge (Beckman Instruments, Palo Alto, Calif., USA) using a Kontrol TZT 45.6 rotor (Kontron AG, Basel, Switzerland). VLDL, IDL and LDL were isolated at densities of 1.006, 1.019 and 1.063 g/ml, respectively. Thereafter, HDL2 and HDL3 were isolated at densities of 1.125 and
349
M.-R. Taskinen et al.: Metabolism of apo A-I and A-II in diabetes
A-I
A-I
H.
~p(A-~/A-I) particles dpPldn' iua~a;i~
Q
Degradation productsin urine Fig. 1. A kinetic model for the metabolism of H D L particles containing apolipoprotein (apo) A-I and A-II (LpA-I/A-II) and apo A-I only (LpA-I). U9 and U~ indicate the inputs of apo A-I into compartments M9 and M1 respectively. Us indicates the input of apo A-II into compartment Ms. EV indicates extravascular space
1.210 g/ml using centrifugation times of 48 h at 38000 rev/min. The concentrations of triglycerides, cholesterol, phospholipids and protein were measured in each fraction. Apo A-I and A-II were assayed from daily samples by immunoturbidometry using monospecific antibodies (nos 726478 and 726486, Boehringer Mannheim GmbH, Mannheim, FRG). The interassay coefficient of variation of apo A-I and A-II assays were 4.3 % and 5.3 % respectively.
Post-heparin plasma LPL and hepatic lipase After collection of the blood sample for H D L isolation, heparin (100 IU/kg body weight, Leiras, Huhtam~iki Oy, Turku, Finland) was injected i. v. as a bolus, and blood was drawn before and 15 rain after the injection into chilled heparinized tubes kept on ice. Plasma was separated immediately in a cooled centrifuge and stored at - 20 ~ LPL and hepatic lipase activities were measured by an immunochemical method using a specific antiserum against hepatic lipase in the LPL assay [25]. Hepatic lipase was measured with a substrate containing 1 tool/1 NaC1 to inactivate the LPL. Each assay series included two reference standards for LPL and HL activities.
Analysis of apo A-I and A-H kinetics Kinetic analysis of the data was performed with the SAAM 20 program [26]. Radioactivity decay curves in plasma and urine, together with the mean masses of apo A-I and A-II were used to derive FCRs and inputs for the individual apoproteins. To gain further insight into apo A kinetics, the turnover data for apo A-I and apo A-II were combined according to the model shown in Figure 1. It is based on the observation that HDL can be divided into populations of particles containing apo A-I and apo A-II (LpA-I/A-II) and apo A-I alone (LpA-I) [22, 27]. The following assumptions were made: (1) Lp A-I and Lp A-I/A-II H D L particles are metabolically distinct and have different input and output rates. ( 2 ) A n HDL particle (LpA-I or LpA-I/A-II) is cleared as a unit with degradation of apoprotein. (3) The molar ratio of A-I to A-II in Lp A-I/A-II is about 1.0.
over was explained by a single plasma compartment while apo A-I required two. The slower metabolised apo A-I pool had kinetic characteristics similar to those of apo A-II. Support for this view of H D L metabolism came from structural studies which revealed that two particle types were present in the HDL density interval, one containing apo A-I the other apo A-I and apo A-II [22, 27, 29, 30]. A series of investigations using immunochemical [31], ultracentrifugal [32] and cross-linking [33] techniques have concluded that the measured molar ratio of apo A-I to apo A-II in LpA-I/A-II is invariant at about 1:1 provided that apo A-II represents a dimer. When measured immunochemically this ratio appears to be approximately 1.3:1, possibly as a result of differential antigenic expression of the apoproteins, giving a weight ratio of 2.1:1. Assumption 3 noted above is, therefore, founded on a substantial body of experimental evidence. ~s~Iapo A-II can be considered to trace the behaviour of LpA-I/ A-II particles assuming the particle is catabolised as a unit (assumption 2) and therefore the calculated FCR for apo A-II provides an FCR for its accompanying apo A-I. This together, the pool size for A-I in LpA-I/A-II derived from the apo A-I/A-II ratio yields an input rate (U9). Derivation of the kinetics of LpA-I is more problematic. There is a difference between apo A-I and A-II clearance as noted above which can be ascribed to differential catabolism of LpA-I and LpAI/A-II particles. If the interparticle exchange of apo A-I is ignored then the initial radioactivity of apo A-I tracer can be distributed according to the calculated mass of apo A-I in LpA-I/A-II and in LpAI (i. e. total apo A-I minus A-I in LpA-I/A-II). The fractional catabolic and input rates of LpA-I can then be derived using SAAM 30 and the model in Figure 1. LpA-I clearance is basically the difference between the behaviour of apo A-I in LpA-I/A-II and that of total apo A-I. If exchange of apo A-I between the two particle types takes place, then the differences between their catabolic rates are reduced and not adequately detected using the present technique. Therefore, the LpA-I kinetic data must be viewed with some caution and may be an underestimate. However, it is useful to apply the model in an attempt to understand more clearly the regulation of H D L metabolism in control and diabetic subjects.
Analytical methods Blood glucose was measured by the glucose oxidase method (Autoanalyser, Technicon, Tarrytown, NY, USA). Serum free-insulin concentrations were determined by radioimmunoassay with Phadeseph insulin radioimmunoassay kits (Pharmacia, Uppsala, Sweden) after precipitation with polyethylene glycol [34]. Serum C-peptide (reference range 0.33-0.67 retool/l) was measured as described by Heding [35]. HbAI (normal range 6-8.5 % ), was measured by microcolumn chromatography (Isolab, Akron, Ohio, USA) [36] and fructosamine (reference range 2.0-2.8 retool/l) by the method of Johnson et al. [37]. The serum concentrations of cholesterol and triglycerides were determined with Boehringer Mannheim kits (nos 18313 and 297771) in a fully automated Olli-D discrete analyser (Kone, Helsinki, Finland).
Statistical analysis The results are presented as mean_+ SD. Data analysis were conducted with a Biomedical Data Processing program [38]. In comparision between the two groups Student's paired t-test (normal distribution) and the Mann-Whitney rank sum test (non-parametric test) were used (program 3D).
Results
Rationale
Metabolic parameters
It has been a long-standing observation that tracer apo A-I is removed from the circulation more rapidly than apo A-II [4-6]. In an attempt to reconcile the behaviour of these major H D L proteins, Zech et al. [28] proposed a composite model in which apo A-II turn-
T h e d i a b e t i c p a t i e n t s e x h i b i t e d a w i d e r a n g e of g l y c a e m i c c o n t r o l ( T a b l e 1). F a s t i n g g l u c o s e c o n c e n t r a t i o n s o v e r t h e 1 3 - d a y t u r n o v e r p e r i o d r a n g e d f r o m 5.4 to 14.5 mmol/1
350
M.-R. Taskinen et al.: Metabolism of apo A-I and A-II in diabetes
Table 1. Plasma lipids and lipoproteins and lipolytic enzyme activities in Type 1 (insulin-dependent)diabetic men (n = 17) and in non-diabetic men (n = 18) Type 1 diabetic patients
Age (years)
BMI (kg/m 2)
Fastingb Triglyceride (mmol/1)c glucose Plasma VLDL (mmol/1)
Cholesterol (mmol/1)c Plasma VLDL LDL
HDL
LPL HL (gmol non-esterified fatty acids .ml-1 .h 1)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 Mean + SD
43 59 43 36 30 33 30 33 37 32 37 23 55 43 25 36 30 37 + 10a
24.6 28.7 26.1 21.5 24.8 21.7 21.6 19.9 21.3 18.1 26.0 24.3 21.3 25.9 22.1 27.4 22.7 23.5 + 2.9
6.8 8.0 11.3 10.0 8.7 12.2 12.3 7.1 6.5 9.8 10.6 8.2 11.9 5.4 14.5 11.9 9.2 9.7 + 2.5
1.01 0.89 0.77 0.97 0.60 1.15 1.33 0.87 0.77 0.81 1.55 1.16 2.02 0.71 0.85 1.00 0.86 1.02+0.35
0.31 0.38 0.29 0.33 0.18 0.46 0.75 0.30 0.34 0.39 0.75 0.61 0.99 0.30 0.39 0.51 0.41 0.45+0.21
6.31 5.25 5.47 6.95 5.52 5.25 6.22 5.25 4.47 3.59 6.24 3.81 8.06 5.87 3.72 4.60 3.39 5.21 _+0.20
0.09 0.18 0.05 0.10 0.07 0.26 0.35 0.16 0.16 0.07 0.88 0.28 0.43 0.11 0.17 0.25 0.08 0.21 + 0.20
4.00 2.90 3.26 4.78 3.59 3.06 4.13 3.36 2.57 1.93 3.72 2.08 6.04 4.41 2.37 3.20 2.30 3.42 + 1.10
2.09 1.99 1.99 1.91 1.82 1.78 1.72 1.71 1.64 1.54 1.34 1.30 1.24 1.24 1.17 0.90 0.84 1.55 + 0.39
52.4 52.9 32.0 20.1 32.6 49.1 22.3 30.0 31.4 36.6 31.3 24.0 27.2 25.8 43.3 15.2 32.0 32.9-+11.0
24.9 28.0 22.8 21.6 19.7 21.4 31.9 23.6 30.4 24.8 29.4 23.5 22.1 64.3 27.5 48.5 35.4 29.4+11.3
20.3 22.8 21.3 27.3 25.2 24.1 26.7 25.8 20.3 25.7 26.9 28.7 26.9 23.2 28.1 24.1 31.1 28.7 25.4 + 3.0
4.7 4.6 4.8 5.0 4.4 4.6 4.7 4.6 4.4 5.3 5.0 5.3 4.8 4.8 4.6 4.6 4.6 4.6 4.7 + 0.3
0.95 0.98 1.30 0.75 0.58 0.73 0.83 1.13 0.83 1.17 2.50 1.12 1.23 0.90 1.80 1.84 1.86 2.36 1.27+0.56
0.43 0.25 0.63 0.18 0.19 0.29 0.28 0.74 0.46 0.65 1.66 0.63 0.75 0.35 1.14 1.28 1.13 1.56 0.70+0.47
5.12 6.35 5.43 6.35 6.26 6.25 5.88 5.41 4.76 4.90 7.00 4.89 4.24 5.99 6.74 5.07 4.84 4.59 5.53+0.83
0.11 0.16 0.60 0.05 0.10 0.09 0.12 0.28 0.18 0.33 0.57 0.23 0.19 1.34 0.53 0.48 0.87 0.82 0.32_+0.26
2.26 3.46 2.22 3.89 3.96 4.20 3.87 3.28 2.89 2.97 4.83 3.05 2.85 4.56 4.86 3.45 2.93 2.97 3.47+0.79
2.66 2.47 2.37 2.27 2.16 1.91 1.76 1.75 1.61 1.53 1.33 1.22 1.i1 1.05 1.04 0.99 0.72 0.51 1.58+0.63
31.1 32.7 39.1 50.1 52.9 27.7 24.9 34.7 12.5 33.6 16.2 16.3
35.1 9.1 12.8 7.9 28.0 27.9 29.8 18.2 45.5 17.3 28.6 29.3
11.4 37.2 7.2 19.2 19.2 29.8+16.1
24.9 70.9 35.6 24.0 19.6 31.7_+21.8
Control subjects 1 53 2 50 3 51 4 51 5 54 6 47 7 49 8 50 9 49 10 50 11 53 12 55 13 25 14 50 15 32 16 55 17 55 18 53 Mean + SD 49 + 8
Post-heparin
ap < 0.01 compared to the control subjects; b The figures represent the mean of the daily measurements over 13 days; c The figures represent the mean of the measurement on days 1, 6, 10 and 13 of the turnover period
( m e a n + SD 9.7 + 2.5 mmol/1). T h e c o n c e n t r a t i o n s of H b A l c a n d f r u c t o s a m i n e a v e r a g e d 8.9 + 1.7 % (range 6.4 to 1 1 . 9 % ) a n d 3.1 +0.54 Bmol/1 (range 2.2 to 3.9 ~tmol/1) respectively.
n o n - d i a b e t i c m e n b u t L D L a n d H D L triglycerides did n o t differ b e t w e e n the groups.
Concentrations of apolipoprotein A-I and A-H Plasma lipids and lipoproteins T h e r e were n o significant differences in the c o n c e n t r a t i o n of cholesterol i n total plasma, V L D L a n d L D L b e t w e e n the two groups. T h e m e a n c o n c e n t r a t i o n s of HDL2 cholesterol a n d HDL3 cholesterol were 1 . 0 3 + 0 . 3 5 a n d 0.52 + 0.14 m m o l / l in Type 1 diabetic patients and 1.06 + 0.56 a n d 0.52 + 0.16 mmol/1 in c o n t r o l m e n respectively. T h e c o n c e n t r a t i o n of triglycerides in total p l a s m a a n d V L D L was lower in Type i diabetic p a t i e n t s t h a n in
M e a n values a n d p l a s m a pools of apo A - I did n o t differ b e t w e e n the diabetic p a t i e n t s a n d the c o n t r o l subjects (Table 2). T h e c o n c e n t r a t i o n of A - I I t e n d e d to b e less in Type i diabetic m e n t h a n in c o n t r o l subjects b u t the diff e r e n c e did n o t reach statistical significance. T h e H D L cholesterol/(apo A - I + apo A - I I ) ratio a n d the apo A - I / a p o A - I I ratio were calculated as indexes of the c o m p o sition of H D L . T h e H D L c h o l e s t e r o l / A - I + A - I I ratio was similar in the two groups (0.36 + 0.05 in Type 1 diabetic subjects vs 0.34 + 0.08 in c o n t r o l subjects, NS) while the
351
M.-R. Taskinen et al.: Metabolism of apo A-I and A-II in diabetes
Table 2. Calculated pool sizes (mg) computed compartment masses (mg), input rates and fractional catabolic rates (FCR) in Type 1 (insulindependent) diabetic men (n = 17) and in non-diabetic men (n = 18) Type 1 diabetic patients
Plasma apo A - F (mg/dl)
Plasma apo A-IF (mg/dl)
FCR (pools/day) apo A-I
apo A-II
apo A-I
Input rates (mg. kg 1. day-1) apo A-II
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 Mean +_SD
158 154 172 166 160 151 134 125 125 127 132 131 140 106 119 98 96 135 _+23
32.2 28.6 34.4 34.1 27.7 42.4 30.1 34.5 24.1 21.2 33.3 31.8 40.0 29.8 24.9 20.4 23.4 30.2 +_6.2
0.116 0.130 0.116 0.138 0.128 0.176 0.154 0.162 0.136 0.157 0.187 0.187 0.187 0.206 0.214 0.277 0.211 0.169 + 0.042
0.130 0.159 0.137 0.119 0.142 0.164 0.149 0.159 0.150 0.167 0.185 0.175 0.190 0.185 0.196 0.235 0.198 0.167 + 0.029
7.32 8.04 8.00 8.82 8.17 10.60 8.71~ 8.11 6.76 7.98 8.60 9.80 10.48 8.75 10.17 10.90 8.07 8.52 +_1.62
1.73 1.82 1.89 1.63 1.57 2.79 1.88 2.18 1.46 1.39 2.47 2.21 3.05 2.19 1.97 1.91 1.84 2.00 + 0.45~
Control subjects ! 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 Mean+ SD
189 212 200 153 167 147 138 146 137 133 116 122 106 93 117 101 109 98 138-+35
34.6 57.5 35.9 30.6 39.1 28.0 38.6 41.1 31.1 38.9 32.4 32.8 26.8 24.1 36.3 30.8 23.4 26.6 33.9+8.0
0.136 0.162 0.158 0.092 0.161 0.130 0.181 0.157 0.163 0.128 0.201 0.185 0.241 0.239 0.235 0.202 0.317 0.324 0.190+0.062
0.151 0.162 0.174 0.104 0.172 0.160 0.162 0.164 0.165 0.142 0.184 0.179 0.230 0.223 0.216 0.201 0.227 0.295 0.184_+0.043
10.29 13.77 13.13 5.36 10.80 7.46 10.00 9.14 8.91 6.83 9.29 9.02 10.26 8.92 11.07 8.10 13.76 12.80 9.45+2.35
2.09 3.73 2.57 1.27 2.68 1.79 2.50 2.70 2.07 2.21 2.40 2.30 2.46 2.13 3.13 2.47 2.17 3.16 2.43_+ 0.55
ap < 0.01 compared to control subjects; b The figures represent the mean of the daily measurements over 13 days
A - I / A - I I r a t i o t e n d e d to b e h i g h e r in the d i a b e t i c m e n (4.55 + 0.78 vs 4.11 + 0.75) t h a n in the c o n t r o l subjects.
Apolipoprotein A-I and A-H metabolism T h e m e a n f r a c t i o n a l c l e a r a n c e for a p o A - I a n d a p o A - I I w e r e n o t significantly d i f f e r e n t b e t w e e n t h e T y p e 1 d i a b e t i c a n d c o n t r o l g r o u p s a l t h o u g h t h e y t e n d e d to b e l o w e r in t h e f o r m e r (Table 2). C a l c u l a t i o n of i n p u t rates showed that Type 1 diabetic men produced the same a m o u n t of a p o A - I as c o n t r o l s u b j e c t s b u t significantly less a p o A - I I . E x a m i n a t i o n of i n d i v i d u a l pairs of a p o A - I a n d a p o A - I I t u r n o v e r s ( T a b l e 2) r e v e a l e d t h a t (1.) t h e F C R for a p o A - I was in s o m e subjects g r e a t e r t h a n a n d in o t h e r s less t h a n t h a t for a p o A - I I , a n d (2.) t h e r e l a t i o n s h i p of a p o A - I a n d a p o A - I I c l e a r a n c e was l i n k e d to t h e r e l a t i v e a b u n d a n c e o f t h e s e a p o p r o t e i n s in p l a s m a . A s s h o w n in F i g u r e 2 w h e n t h e p l a s m a a p o A - I level was low, its clearance e x c e e d e d t h a t o f a p o A - I I while t h e c o n v e r s e was
t r u e at high a p o A - I levels. N e a r the p o p u l a t i o n m e a n v a l u e for t h e p l a s m a c o n c e n t r a t i o n of t h e s e a p o p r o t e i n s ( m i d d l e p a n e l ) , t h e d e c a y rates for A - I a n d A - I I w e r e virt u a l l y identical. T h e s e o b s e r v a t i o n s c a n n o t b e e x p l a i n e d unless the t r a c e r s a r e e n v i s a g e d as b e i n g u n e q u a l l y dist r i b u t e d w i t h i n t h e H D L s p e c t r u m as c o n c e i v e d in t h e m o d e l in F i g u r e 1. If it is a c c e p t e d t h a t ~31I a p o A - I I traces t h e L p A - I / A - I I p a r t i c l e a n d the L p A - I b e h a v e s i n d e p e n d e n t l y t h e n it is p o s s i b l e to d e r i v e a d d i t i o n a l k i n e t i c p a r a m e t e r s for t h e s e p a r t i c l e s as s h o w n in T a b l e 3. W h i l e L p A - I / A - I I c l e a r a n c e is r e l a t i v e l y c o n s t a n t in b o t h g r o u p s t h e r a t e o f L p A - I c a t a b o l i s m is m o r e v a r i a b l e a n d m a y b e less or g r e a t e r t h a n t h a t of L p A - I / A - I I . Subjects with a higher total plasma apo A-I concentration have higher levels of L p A - I a n d a s l o w e r c l e a r a n c e o f this p a r t i c l e (Table 3). W h e n t h e d a t a f r o m all 35 s u b j e c t s w e r e c o m b i n e d ( F i g . 3 ) a n d d i v i d e d in tertiles of L p A - I F C R , it c o u l d b e s e e n t h a t t h o s e in t h e l o w e s t tertile h a d high conc e n t r a t i o n s of H D L 2 c h o l e s t e r o l a n d p l a s m a a p o A - I a n d r e l a t i v e l y l a r g e r H D L ( b a s e d on t h e H D L c h o l e s t e r o l /
352
M.-R. Taskinen et al.: Metabolism of apo A-I and A-II in diabetes
1O0
was examined. In non-diabetic men a significant positive correlation existed between the apo A - I and apo A - I I F C R and V L D L triglyceride concentration (Fig.5) but this was absent in the Type 1 diabetic group. In contrast, a positive relationship was seen between V L D L triglyceride and apo A - I or apo A - I I input rates ( m g . k g 1. d a y - l) (r = + 0.49,p < 0.05, r = + 0.67,p < 0.01) in Type 1 diabetic men but non-diabetic in normal m e n (r = 0.24, r = 0.29, NS).
2(
1001"%
~~ 5
"-
.
100
Lipolytic enzymes, HD L and apo A metabolism 2~
10
2
4
6
8
10
12
lP4
Days
Fig.2. Radioactive decay curves of ~25I-labelled apolipoprotein (apo) A-I and mI-labelled apo A-II in three individuals with either high (apo A-I = 156 mg/dl and apo A-II = 29 mg/dl, upper panel), moderate (apo A-I = 137 mg/dl and apo A-II = 31 mg/dl, middle panel) or low (apo A-I = 109 mg/dl and apo A-II = 24 mg/dl, lower panel) plasma concentration of apo A-I
A-I + A - I I ratio). Mean values for apo A input rate for the Type 1 diabetic patients and control subjects are given in Table 3. The input of apo A - I and L p A - I particles (U9, Table 3, Figs. 1 and 4) was similar in the two groups. In contrast, the input rates for apo A - I (U1, Figs. 1 and 4) and apo A - I I (Us, Figs. 1 and 4) into L p A - I / A - I I were reduced in Type 1 diabetic patients.
Interrelations between apo A metabolism and plasma H D L and V L D L levels H D L cholesterol correlated inversely with FCRs of apo A-I, apo A - I I (r = - 0 . 9 1 in Type i diabetic patients, r = - 0 . 8 3 in control subjects; p < 0.001 in both groups) and the calculated F C R for L p A - I (r = - 0.08 in Type 1 diabetic patients, r = - 0.72 in control subjects, p < 0.001 in both groups). These correlations were accounted for by the variation of HDL2 but not HDL3. T h e r e was no association between H D L cholesterol levels and any of the input rates for apo A - I or apo A - I I in non-diabetic or Type i diabetic men (Tables 2 and 3). The concentration of apo A - I was determined by the rate of catabolism of the protein (plasma apo A - I vs apo A - I FCR, r = -0.82, p
Diabetologia 9 Springer-Verlag 1992
Metabolism of HDL apolipoprotein A-I and A-II in Type 1 (insulin-dependent) diabetes mellitus M.-R. T a s k i n e n 1, J. Kahri 1, V. K o i v i s t o 2, J. Shepherd 3 and C.J. Packard 3
1Third and 2Second Departments of Medicine, University of Helsinki, Finland and the 3Institute of Biochemistry, Royal Infirmary, Glasgow, Scotland, UK
Summary. Concentrations of HDL cholesterol and apolipoprotein A-I are commonly increased in Type 1 (insulindependent) diabetes mellitus but the mechanisms whereby diabetes influences H D L metabolism have not been studied. We investigated the metabolism of HDL apoproteins A-I and II in normolipidaemic Type 1 diabetic men (n = 17, HbA~ 6.4-11.9 % ) without microalbuminuria but with a wide range of H D L cholesterol (0.85-2.10 mmol/1) and in nondiabetic men (n = 18) matched for body mass index and the range of H D L cholesterol. Input rates and fractional catabolic rates for apolipoproteins A-I and II were determined following injection of ~2q-apolipoprotein A-I and ~31I-apolipoprotein A-II tracers. Additional multicompartmental analysis was performed using a model to describe the kinetics of HDL particles containing only apolipoprotein A-I (Lp A-I) and apolipoprotein A-I and apolipoprotein A-II (Lp A-I/ A-II). No gross differences from normal subjects were observed in the mean levels of lipids, lipoproteins, apoproteins and the lipolytic enzymes in the diabetic men as a result of the selection process. Furthermore, the relationship between
H D L cholesterol, apolipoprotein A-I, apolipoprotein A-II, kinetic analyses, VLDL triglyceride, lipolytic enzymes.
A reduced H D L cholesterol level in plasma is a recognised risk factor for coronary heart disease [1]. The concentration and composition of this lipoprotein are regulated by the combined actions of a number of factors including lipoprotein lipase (LPL), hepatic lipase (HL) and cholesteryl-ester transfer protein (CETP) [2, 3]. In addition plasma H D L levels are known to be influenced by changes in the synthesis and catabolism of the major H D L proteins, apolipoproteins (apo) A-I and A-II. H D L turnover studies have shown that the fractional catabolic rate (FCR) of apo A-I correlates inversely with the H D L levels in the circulation [4-7], i. e. a slow apo A-I F C R is associated with a high concentration of apo A-I and H D L . In contrast, fast clearance of apo A-I leads to low H D L levels regardless of the presence or absence of hypertriglyceridaemia [7-9]. Since coronary heart disease is the principal complication and leading cause of death in diabetes, H D L has been
a focus of great interest in both the insulin-dependent and non-insulin-dependent forms of the condition. In general H D L cholesterol is lower in Type 2 (non-insulin-dependent) patients compared to non-diabetic populations matched for sex, age and body mass index [10]. Conversely, most studies have reported that in Type i diabetic patients with fair to good glycaemic control, the concentration of H D L cholesterol is either increased or normal [10]. Lowering of H D L cholesterol has been documented in two studies [11, 12]. Thus, the response of H D L in diabetes seems to be variable. The observed elevation of H D L cholesterol in Type i diabetes is mainly due to high levels of HDL2 but a rise of HDL3 has also been reported [13-16]. Not unexpectedly, therefore, the concentration of apo A-I in the plasma is generally elevated whereas that of apo A-II seems to be within the normal range [17-19]. Hyper-apo A-I is considered to be anti-atherogenic [3]. The apo A-I/A-II ratio rises consistently with a preferen-
apolipoprotein A kinetics and plasma HDL cholesterol levels appeared to be preserved in the diabetic group. However, some normal interrelationships were disrupted in the diabetic men. Firstly, the rate of apolipoprotein A-II synthesis was 22% lower than in control subjects (p < 0.05). Modelling indicated that this was due to decreased input of Lp A-I/A-II particles whereas the input of Lp A-I particles was similar in the two groups. Secondly, there was no correlation between VLDL triglyceride and H D L cholesterol or VLDL triglyceride and the fractional catabolic rate of apolipoproteins A-I and A-II in diabetic men in contrast to that seen in control subjects. We conclude that there is a disruption in the normal association between VLDL and HDL metabolism in Type i diabetic men and postulate that the observed differences may be due to the therapeutic use of exogenous insulin. Key words: Type 1 (insulin-dependent) diabetes mellitus,
348 tial e l e v a t i o n of HDL2 which is the a n t i - a t h e r o g e n i c subclass of H D L cholesterol [3, 10]. C o n s e q u e n t l y there is an u n s o l v e d p a r a d o x in Type 1 diabetes which is in g e n e r a l associated with a n increased risk of c o r o n a r y h e a r t disease. O v e r a l l the m e c h a n i s m s by which diabetes m o d i f y H D L m e t a b o l i s m are n o t fully u n d e r s t o o d . We have s h o w n that in Type i diabetes, as in the n o n - d i a b e t i c p o p u l a t i o n , lipolytic e n z y m e s ( L P L a n d H L ) regulate p l a s m a HDL2 [14]. T h e o b s e r v e d e l e v a t i o n of HDL2 cholesterol c a n be at least partly e x p l a i n e d b y a rise of L P L activity due to p e r i p h e r a l h y p e r i n s u l i n i s m in c o n v e n tionally t r e a t e d Type 1 diabetes [20]. However, there are n o reports o n the m e t a b o l i s m of apo A - I a n d A - I I in Type 1 diabetic patients. I n the p r e s e n t study we have e x a m i n e d in vivo the t u r n over of apo A - I a n d A - I I in n o r m o l i p i d a e m i c Type 1 diabetic subjects with fair to good glycaemic control a n d a wide r a n g e of H D L cholesterol. I n particular, we sought to d e t e r m i n e w h e t h e r the diabetic c o n d i t i o n per se inf l u e n c e d H D L m e t a b o l i s m in Type 1 diabetes. Consequently, the p a t i e n t s were c o m p a r e d to n o n - d i a b e t i c control subjects with a similar wide r a n g e of H D L levels. I n addition, we m e a s u r e d the activities of lipolytic e n z y m e s ( L P L a n d H L ) in p o s t - h e p a r i n i z e d plasma.
Subjects and methods Seventeen Type i diabetic men selected to provide a wide range of HDL cholesterol (0.85-2.10 retool/l) but with normal serum triglyceride concentrations ( < 2.5 mmol/1) participated in the study. The average age of the patients was 36_+2 years (mean + SEM, range 23-59 years) and the mean body mass index ranged from 18.1 to 27.4 kg/m 2(mean _+SEM, 23.5 + 0.6 kg/m2). The duration of diabetes averaged 10.6 + 1.8 years (range 2-25 years). Sixteen patients received insulin by injection, two (n = 5) to three (n = 11) times daily and in one patient it was administered subcutaneously via a pump. The average dose of daily insulinwas 41 + 3 IU which remained constant throughout the study. The concentrations of HbAl and fructosamine ranged from 6.4 to 11.9% and 2.2 to 3.9 gmol/1 respectively. All the patients had C-peptide values of less than 0.33 nmol/1. Overnight albumin excretion rates (3-18 gg/min) were normal as was the mean serum creatinine at 94 + 11 (mean + SEM) gmol/1. Two patients had background retinopathy. None was taking lipid-lowering drugs or other drugs known to influence lipid metabolism. The patients were instructed to follow a weight-maintaining, sucrose-free diet containing 45 % of its calories as carbohydrate, 35 % as fat and 20% as protein. The patients were recruited from the Diabetes Clinic of the Helsinki University Central Hospital. Control men (n = 18) were recruited from volunteers screened for the Helsinki Heart Study between 1982 and 1983 but excluded from the study because their non-HDL cholesterol was < 5.2 mmol/1. A subgroup (n = 92) of these men living in the Helsinki area were screened again in 1988. The control group was selected to have similar ranges of HDL cholesterol and serum triglyceride as the Type 1 diabetic patients. None had a history of cardiovascular disease. Endocrine and other disorders which could influence lipoprotein metabolism were excluded by medical history, clinical examination and liver, kidney and thyroid function tests. None of the subjects was taking any medication. The absence of diabetes was confirmed by measurement of fasting blood glucose and serum fructosamine. All clinical and laboratory work was cornpleted in Helsinki whereas the kinetic analyses of apo A-I and A-II were performed in Glasgow. The study protocol was approved by the Ethical Committee of Helsinki University Hospital and informed consent was obtained from each subject.
M.-R. Taskinen et al.: Metabolism of apo A-I and A-II in diabetes
Purification and labelling of apolipoprotein A-I and A-H Heterologous apo A-I and A-II were used to guarantee the same batch and structure of apoprotein. Apo A-I and A-II were prepared in Glasgow from the plasma o f healthy donors, who were negative on screening for hepatitis B and human immunodeficiencyvirus. HDL was separated by ultracentrifugationand apo A-I and A-II prepared after delipidation and purification using high performance liquid chromatography as described previously [21]. The purity of apoproteins was established by a battery of immunoassays and by amino acid analysis [22]. The isolated apoproteins were dialysed against 0.1mot/I NH4HCO3 (pH 8.6), lyophilized and stored at -70~ Batches of lyophilized apo A-I and A-II were mailed to Helsinki in dry ice and stored at -70~ Apo A-I (0.5mg) and apo A-II (0.5 mg) were trace labelled with i mCi of 1~3Iand I mCi 131Irespectively, by the iodine monochloride method [21, 23]. Radioiodinated apoproteins were freed from unbound radioiodide by column chromatography (Sephadex G-25M, Pharmacia, Uppsala, Sweden) and used immediately.
Kinetic studies The subjects were examined as out-patients to avoid disruption of their daily life. No alcohol intake was allowed during the study and the subjects were instructed to maintain their normal physical activity. For 3 days before and throughout the study the subjects received potassium iodide (60 rag, three times per day) to minimise thyroid absorption of radioactive iodide. Fasting blood was drawn into sterile tubes containing 300 B1 of 0.40 mol/l disodium-EDTA solution in 80ml samples. Plasma (25 ml) was adjusted to a density of 1.063 g/ml by adding solid KBr and centrifuged in a Beckman Ti 60 rotor (24 h, 40000 rev/min, 4 ~ The top fraction containingVLDL, intermediate density Iipoprotein (IDL) and LDL was removed and the density of the bottom adjusted to 1.210 g/ml by addition of solid KBr. HDL was isolated by a further centrifugation for 24 h as above. The HDL fraction was harvested by aspiration of 1-2 ml samples and stored at 4 ~ for up to 1-2 h. 125I-labelledapo A-I and 131I-labelledapo A-II were incubated with 0.5-1 ml of the HDL of each subject for 30 rain at room temperature. The density of the incubation mixtures was adjusted to 1.21 g/mI by using solid KBr, overlayered with 3.5 ml of KBr solution (density = 1.210 g/ml) and the HDL re-isolated by ultracentrifugation for 21 h at 40000 rev/min at 4~ in a Beckman Ti 50.3 rotor. ~25Iapo A-I HDL and 131Iapo A-II HDL were harvested by aspiration and dialysed extensively against sterile 0.15 mol/1 NaC1, 0.01% disodium-EDTA. Thereafter the preparations were sterilised by membrane filtration (0.22 gmol Millipore filters, Millipore, Bedford, Mass., USA). On the fifth morning after the collection of the initial blood samples the subjects received 25 gCi each of labelled apo A-I HDL and apo A-II HDL by i.v. bolus injection. The first blood sample was taken after 10 rain and then daily after a 10-h overnight fast until day 13.24-h urine specimens were collected over 13 days and the urine volume measured. The concentration of creatinine in the urine was used as an index of completeness of collection. The radioactivity was measured in 2-ml aliquots of plasma and urine, and was counted at the end of the study.
Lipid, lipoprotein and apoprotein determinations On days 1,6, 10 and 13 of the turnover period blood was drawn after a 10-h overnight fast for measurement of serum lipids and lipoproteins. Lipoprotein fractions were isolated by sequential ultracentrifugation [24] in a Beckman L7-70 ultracentrifuge (Beckman Instruments, Palo Alto, Calif., USA) using a Kontrol TZT 45.6 rotor (Kontron AG, Basel, Switzerland). VLDL, IDL and LDL were isolated at densities of 1.006, 1.019 and 1.063 g/ml, respectively. Thereafter, HDL2 and HDL3 were isolated at densities of 1.125 and
349
M.-R. Taskinen et al.: Metabolism of apo A-I and A-II in diabetes
A-I
A-I
H.
~p(A-~/A-I) particles dpPldn' iua~a;i~
Q
Degradation productsin urine Fig. 1. A kinetic model for the metabolism of H D L particles containing apolipoprotein (apo) A-I and A-II (LpA-I/A-II) and apo A-I only (LpA-I). U9 and U~ indicate the inputs of apo A-I into compartments M9 and M1 respectively. Us indicates the input of apo A-II into compartment Ms. EV indicates extravascular space
1.210 g/ml using centrifugation times of 48 h at 38000 rev/min. The concentrations of triglycerides, cholesterol, phospholipids and protein were measured in each fraction. Apo A-I and A-II were assayed from daily samples by immunoturbidometry using monospecific antibodies (nos 726478 and 726486, Boehringer Mannheim GmbH, Mannheim, FRG). The interassay coefficient of variation of apo A-I and A-II assays were 4.3 % and 5.3 % respectively.
Post-heparin plasma LPL and hepatic lipase After collection of the blood sample for H D L isolation, heparin (100 IU/kg body weight, Leiras, Huhtam~iki Oy, Turku, Finland) was injected i. v. as a bolus, and blood was drawn before and 15 rain after the injection into chilled heparinized tubes kept on ice. Plasma was separated immediately in a cooled centrifuge and stored at - 20 ~ LPL and hepatic lipase activities were measured by an immunochemical method using a specific antiserum against hepatic lipase in the LPL assay [25]. Hepatic lipase was measured with a substrate containing 1 tool/1 NaC1 to inactivate the LPL. Each assay series included two reference standards for LPL and HL activities.
Analysis of apo A-I and A-H kinetics Kinetic analysis of the data was performed with the SAAM 20 program [26]. Radioactivity decay curves in plasma and urine, together with the mean masses of apo A-I and A-II were used to derive FCRs and inputs for the individual apoproteins. To gain further insight into apo A kinetics, the turnover data for apo A-I and apo A-II were combined according to the model shown in Figure 1. It is based on the observation that HDL can be divided into populations of particles containing apo A-I and apo A-II (LpA-I/A-II) and apo A-I alone (LpA-I) [22, 27]. The following assumptions were made: (1) Lp A-I and Lp A-I/A-II H D L particles are metabolically distinct and have different input and output rates. ( 2 ) A n HDL particle (LpA-I or LpA-I/A-II) is cleared as a unit with degradation of apoprotein. (3) The molar ratio of A-I to A-II in Lp A-I/A-II is about 1.0.
over was explained by a single plasma compartment while apo A-I required two. The slower metabolised apo A-I pool had kinetic characteristics similar to those of apo A-II. Support for this view of H D L metabolism came from structural studies which revealed that two particle types were present in the HDL density interval, one containing apo A-I the other apo A-I and apo A-II [22, 27, 29, 30]. A series of investigations using immunochemical [31], ultracentrifugal [32] and cross-linking [33] techniques have concluded that the measured molar ratio of apo A-I to apo A-II in LpA-I/A-II is invariant at about 1:1 provided that apo A-II represents a dimer. When measured immunochemically this ratio appears to be approximately 1.3:1, possibly as a result of differential antigenic expression of the apoproteins, giving a weight ratio of 2.1:1. Assumption 3 noted above is, therefore, founded on a substantial body of experimental evidence. ~s~Iapo A-II can be considered to trace the behaviour of LpA-I/ A-II particles assuming the particle is catabolised as a unit (assumption 2) and therefore the calculated FCR for apo A-II provides an FCR for its accompanying apo A-I. This together, the pool size for A-I in LpA-I/A-II derived from the apo A-I/A-II ratio yields an input rate (U9). Derivation of the kinetics of LpA-I is more problematic. There is a difference between apo A-I and A-II clearance as noted above which can be ascribed to differential catabolism of LpA-I and LpAI/A-II particles. If the interparticle exchange of apo A-I is ignored then the initial radioactivity of apo A-I tracer can be distributed according to the calculated mass of apo A-I in LpA-I/A-II and in LpAI (i. e. total apo A-I minus A-I in LpA-I/A-II). The fractional catabolic and input rates of LpA-I can then be derived using SAAM 30 and the model in Figure 1. LpA-I clearance is basically the difference between the behaviour of apo A-I in LpA-I/A-II and that of total apo A-I. If exchange of apo A-I between the two particle types takes place, then the differences between their catabolic rates are reduced and not adequately detected using the present technique. Therefore, the LpA-I kinetic data must be viewed with some caution and may be an underestimate. However, it is useful to apply the model in an attempt to understand more clearly the regulation of H D L metabolism in control and diabetic subjects.
Analytical methods Blood glucose was measured by the glucose oxidase method (Autoanalyser, Technicon, Tarrytown, NY, USA). Serum free-insulin concentrations were determined by radioimmunoassay with Phadeseph insulin radioimmunoassay kits (Pharmacia, Uppsala, Sweden) after precipitation with polyethylene glycol [34]. Serum C-peptide (reference range 0.33-0.67 retool/l) was measured as described by Heding [35]. HbAI (normal range 6-8.5 % ), was measured by microcolumn chromatography (Isolab, Akron, Ohio, USA) [36] and fructosamine (reference range 2.0-2.8 retool/l) by the method of Johnson et al. [37]. The serum concentrations of cholesterol and triglycerides were determined with Boehringer Mannheim kits (nos 18313 and 297771) in a fully automated Olli-D discrete analyser (Kone, Helsinki, Finland).
Statistical analysis The results are presented as mean_+ SD. Data analysis were conducted with a Biomedical Data Processing program [38]. In comparision between the two groups Student's paired t-test (normal distribution) and the Mann-Whitney rank sum test (non-parametric test) were used (program 3D).
Results
Rationale
Metabolic parameters
It has been a long-standing observation that tracer apo A-I is removed from the circulation more rapidly than apo A-II [4-6]. In an attempt to reconcile the behaviour of these major H D L proteins, Zech et al. [28] proposed a composite model in which apo A-II turn-
T h e d i a b e t i c p a t i e n t s e x h i b i t e d a w i d e r a n g e of g l y c a e m i c c o n t r o l ( T a b l e 1). F a s t i n g g l u c o s e c o n c e n t r a t i o n s o v e r t h e 1 3 - d a y t u r n o v e r p e r i o d r a n g e d f r o m 5.4 to 14.5 mmol/1
350
M.-R. Taskinen et al.: Metabolism of apo A-I and A-II in diabetes
Table 1. Plasma lipids and lipoproteins and lipolytic enzyme activities in Type 1 (insulin-dependent)diabetic men (n = 17) and in non-diabetic men (n = 18) Type 1 diabetic patients
Age (years)
BMI (kg/m 2)
Fastingb Triglyceride (mmol/1)c glucose Plasma VLDL (mmol/1)
Cholesterol (mmol/1)c Plasma VLDL LDL
HDL
LPL HL (gmol non-esterified fatty acids .ml-1 .h 1)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 Mean + SD
43 59 43 36 30 33 30 33 37 32 37 23 55 43 25 36 30 37 + 10a
24.6 28.7 26.1 21.5 24.8 21.7 21.6 19.9 21.3 18.1 26.0 24.3 21.3 25.9 22.1 27.4 22.7 23.5 + 2.9
6.8 8.0 11.3 10.0 8.7 12.2 12.3 7.1 6.5 9.8 10.6 8.2 11.9 5.4 14.5 11.9 9.2 9.7 + 2.5
1.01 0.89 0.77 0.97 0.60 1.15 1.33 0.87 0.77 0.81 1.55 1.16 2.02 0.71 0.85 1.00 0.86 1.02+0.35
0.31 0.38 0.29 0.33 0.18 0.46 0.75 0.30 0.34 0.39 0.75 0.61 0.99 0.30 0.39 0.51 0.41 0.45+0.21
6.31 5.25 5.47 6.95 5.52 5.25 6.22 5.25 4.47 3.59 6.24 3.81 8.06 5.87 3.72 4.60 3.39 5.21 _+0.20
0.09 0.18 0.05 0.10 0.07 0.26 0.35 0.16 0.16 0.07 0.88 0.28 0.43 0.11 0.17 0.25 0.08 0.21 + 0.20
4.00 2.90 3.26 4.78 3.59 3.06 4.13 3.36 2.57 1.93 3.72 2.08 6.04 4.41 2.37 3.20 2.30 3.42 + 1.10
2.09 1.99 1.99 1.91 1.82 1.78 1.72 1.71 1.64 1.54 1.34 1.30 1.24 1.24 1.17 0.90 0.84 1.55 + 0.39
52.4 52.9 32.0 20.1 32.6 49.1 22.3 30.0 31.4 36.6 31.3 24.0 27.2 25.8 43.3 15.2 32.0 32.9-+11.0
24.9 28.0 22.8 21.6 19.7 21.4 31.9 23.6 30.4 24.8 29.4 23.5 22.1 64.3 27.5 48.5 35.4 29.4+11.3
20.3 22.8 21.3 27.3 25.2 24.1 26.7 25.8 20.3 25.7 26.9 28.7 26.9 23.2 28.1 24.1 31.1 28.7 25.4 + 3.0
4.7 4.6 4.8 5.0 4.4 4.6 4.7 4.6 4.4 5.3 5.0 5.3 4.8 4.8 4.6 4.6 4.6 4.6 4.7 + 0.3
0.95 0.98 1.30 0.75 0.58 0.73 0.83 1.13 0.83 1.17 2.50 1.12 1.23 0.90 1.80 1.84 1.86 2.36 1.27+0.56
0.43 0.25 0.63 0.18 0.19 0.29 0.28 0.74 0.46 0.65 1.66 0.63 0.75 0.35 1.14 1.28 1.13 1.56 0.70+0.47
5.12 6.35 5.43 6.35 6.26 6.25 5.88 5.41 4.76 4.90 7.00 4.89 4.24 5.99 6.74 5.07 4.84 4.59 5.53+0.83
0.11 0.16 0.60 0.05 0.10 0.09 0.12 0.28 0.18 0.33 0.57 0.23 0.19 1.34 0.53 0.48 0.87 0.82 0.32_+0.26
2.26 3.46 2.22 3.89 3.96 4.20 3.87 3.28 2.89 2.97 4.83 3.05 2.85 4.56 4.86 3.45 2.93 2.97 3.47+0.79
2.66 2.47 2.37 2.27 2.16 1.91 1.76 1.75 1.61 1.53 1.33 1.22 1.i1 1.05 1.04 0.99 0.72 0.51 1.58+0.63
31.1 32.7 39.1 50.1 52.9 27.7 24.9 34.7 12.5 33.6 16.2 16.3
35.1 9.1 12.8 7.9 28.0 27.9 29.8 18.2 45.5 17.3 28.6 29.3
11.4 37.2 7.2 19.2 19.2 29.8+16.1
24.9 70.9 35.6 24.0 19.6 31.7_+21.8
Control subjects 1 53 2 50 3 51 4 51 5 54 6 47 7 49 8 50 9 49 10 50 11 53 12 55 13 25 14 50 15 32 16 55 17 55 18 53 Mean + SD 49 + 8
Post-heparin
ap < 0.01 compared to the control subjects; b The figures represent the mean of the daily measurements over 13 days; c The figures represent the mean of the measurement on days 1, 6, 10 and 13 of the turnover period
( m e a n + SD 9.7 + 2.5 mmol/1). T h e c o n c e n t r a t i o n s of H b A l c a n d f r u c t o s a m i n e a v e r a g e d 8.9 + 1.7 % (range 6.4 to 1 1 . 9 % ) a n d 3.1 +0.54 Bmol/1 (range 2.2 to 3.9 ~tmol/1) respectively.
n o n - d i a b e t i c m e n b u t L D L a n d H D L triglycerides did n o t differ b e t w e e n the groups.
Concentrations of apolipoprotein A-I and A-H Plasma lipids and lipoproteins T h e r e were n o significant differences in the c o n c e n t r a t i o n of cholesterol i n total plasma, V L D L a n d L D L b e t w e e n the two groups. T h e m e a n c o n c e n t r a t i o n s of HDL2 cholesterol a n d HDL3 cholesterol were 1 . 0 3 + 0 . 3 5 a n d 0.52 + 0.14 m m o l / l in Type 1 diabetic patients and 1.06 + 0.56 a n d 0.52 + 0.16 mmol/1 in c o n t r o l m e n respectively. T h e c o n c e n t r a t i o n of triglycerides in total p l a s m a a n d V L D L was lower in Type i diabetic p a t i e n t s t h a n in
M e a n values a n d p l a s m a pools of apo A - I did n o t differ b e t w e e n the diabetic p a t i e n t s a n d the c o n t r o l subjects (Table 2). T h e c o n c e n t r a t i o n of A - I I t e n d e d to b e less in Type i diabetic m e n t h a n in c o n t r o l subjects b u t the diff e r e n c e did n o t reach statistical significance. T h e H D L cholesterol/(apo A - I + apo A - I I ) ratio a n d the apo A - I / a p o A - I I ratio were calculated as indexes of the c o m p o sition of H D L . T h e H D L c h o l e s t e r o l / A - I + A - I I ratio was similar in the two groups (0.36 + 0.05 in Type 1 diabetic subjects vs 0.34 + 0.08 in c o n t r o l subjects, NS) while the
351
M.-R. Taskinen et al.: Metabolism of apo A-I and A-II in diabetes
Table 2. Calculated pool sizes (mg) computed compartment masses (mg), input rates and fractional catabolic rates (FCR) in Type 1 (insulindependent) diabetic men (n = 17) and in non-diabetic men (n = 18) Type 1 diabetic patients
Plasma apo A - F (mg/dl)
Plasma apo A-IF (mg/dl)
FCR (pools/day) apo A-I
apo A-II
apo A-I
Input rates (mg. kg 1. day-1) apo A-II
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 Mean +_SD
158 154 172 166 160 151 134 125 125 127 132 131 140 106 119 98 96 135 _+23
32.2 28.6 34.4 34.1 27.7 42.4 30.1 34.5 24.1 21.2 33.3 31.8 40.0 29.8 24.9 20.4 23.4 30.2 +_6.2
0.116 0.130 0.116 0.138 0.128 0.176 0.154 0.162 0.136 0.157 0.187 0.187 0.187 0.206 0.214 0.277 0.211 0.169 + 0.042
0.130 0.159 0.137 0.119 0.142 0.164 0.149 0.159 0.150 0.167 0.185 0.175 0.190 0.185 0.196 0.235 0.198 0.167 + 0.029
7.32 8.04 8.00 8.82 8.17 10.60 8.71~ 8.11 6.76 7.98 8.60 9.80 10.48 8.75 10.17 10.90 8.07 8.52 +_1.62
1.73 1.82 1.89 1.63 1.57 2.79 1.88 2.18 1.46 1.39 2.47 2.21 3.05 2.19 1.97 1.91 1.84 2.00 + 0.45~
Control subjects ! 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 Mean+ SD
189 212 200 153 167 147 138 146 137 133 116 122 106 93 117 101 109 98 138-+35
34.6 57.5 35.9 30.6 39.1 28.0 38.6 41.1 31.1 38.9 32.4 32.8 26.8 24.1 36.3 30.8 23.4 26.6 33.9+8.0
0.136 0.162 0.158 0.092 0.161 0.130 0.181 0.157 0.163 0.128 0.201 0.185 0.241 0.239 0.235 0.202 0.317 0.324 0.190+0.062
0.151 0.162 0.174 0.104 0.172 0.160 0.162 0.164 0.165 0.142 0.184 0.179 0.230 0.223 0.216 0.201 0.227 0.295 0.184_+0.043
10.29 13.77 13.13 5.36 10.80 7.46 10.00 9.14 8.91 6.83 9.29 9.02 10.26 8.92 11.07 8.10 13.76 12.80 9.45+2.35
2.09 3.73 2.57 1.27 2.68 1.79 2.50 2.70 2.07 2.21 2.40 2.30 2.46 2.13 3.13 2.47 2.17 3.16 2.43_+ 0.55
ap < 0.01 compared to control subjects; b The figures represent the mean of the daily measurements over 13 days
A - I / A - I I r a t i o t e n d e d to b e h i g h e r in the d i a b e t i c m e n (4.55 + 0.78 vs 4.11 + 0.75) t h a n in the c o n t r o l subjects.
Apolipoprotein A-I and A-H metabolism T h e m e a n f r a c t i o n a l c l e a r a n c e for a p o A - I a n d a p o A - I I w e r e n o t significantly d i f f e r e n t b e t w e e n t h e T y p e 1 d i a b e t i c a n d c o n t r o l g r o u p s a l t h o u g h t h e y t e n d e d to b e l o w e r in t h e f o r m e r (Table 2). C a l c u l a t i o n of i n p u t rates showed that Type 1 diabetic men produced the same a m o u n t of a p o A - I as c o n t r o l s u b j e c t s b u t significantly less a p o A - I I . E x a m i n a t i o n of i n d i v i d u a l pairs of a p o A - I a n d a p o A - I I t u r n o v e r s ( T a b l e 2) r e v e a l e d t h a t (1.) t h e F C R for a p o A - I was in s o m e subjects g r e a t e r t h a n a n d in o t h e r s less t h a n t h a t for a p o A - I I , a n d (2.) t h e r e l a t i o n s h i p of a p o A - I a n d a p o A - I I c l e a r a n c e was l i n k e d to t h e r e l a t i v e a b u n d a n c e o f t h e s e a p o p r o t e i n s in p l a s m a . A s s h o w n in F i g u r e 2 w h e n t h e p l a s m a a p o A - I level was low, its clearance e x c e e d e d t h a t o f a p o A - I I while t h e c o n v e r s e was
t r u e at high a p o A - I levels. N e a r the p o p u l a t i o n m e a n v a l u e for t h e p l a s m a c o n c e n t r a t i o n of t h e s e a p o p r o t e i n s ( m i d d l e p a n e l ) , t h e d e c a y rates for A - I a n d A - I I w e r e virt u a l l y identical. T h e s e o b s e r v a t i o n s c a n n o t b e e x p l a i n e d unless the t r a c e r s a r e e n v i s a g e d as b e i n g u n e q u a l l y dist r i b u t e d w i t h i n t h e H D L s p e c t r u m as c o n c e i v e d in t h e m o d e l in F i g u r e 1. If it is a c c e p t e d t h a t ~31I a p o A - I I traces t h e L p A - I / A - I I p a r t i c l e a n d the L p A - I b e h a v e s i n d e p e n d e n t l y t h e n it is p o s s i b l e to d e r i v e a d d i t i o n a l k i n e t i c p a r a m e t e r s for t h e s e p a r t i c l e s as s h o w n in T a b l e 3. W h i l e L p A - I / A - I I c l e a r a n c e is r e l a t i v e l y c o n s t a n t in b o t h g r o u p s t h e r a t e o f L p A - I c a t a b o l i s m is m o r e v a r i a b l e a n d m a y b e less or g r e a t e r t h a n t h a t of L p A - I / A - I I . Subjects with a higher total plasma apo A-I concentration have higher levels of L p A - I a n d a s l o w e r c l e a r a n c e o f this p a r t i c l e (Table 3). W h e n t h e d a t a f r o m all 35 s u b j e c t s w e r e c o m b i n e d ( F i g . 3 ) a n d d i v i d e d in tertiles of L p A - I F C R , it c o u l d b e s e e n t h a t t h o s e in t h e l o w e s t tertile h a d high conc e n t r a t i o n s of H D L 2 c h o l e s t e r o l a n d p l a s m a a p o A - I a n d r e l a t i v e l y l a r g e r H D L ( b a s e d on t h e H D L c h o l e s t e r o l /
352
M.-R. Taskinen et al.: Metabolism of apo A-I and A-II in diabetes
1O0
was examined. In non-diabetic men a significant positive correlation existed between the apo A - I and apo A - I I F C R and V L D L triglyceride concentration (Fig.5) but this was absent in the Type 1 diabetic group. In contrast, a positive relationship was seen between V L D L triglyceride and apo A - I or apo A - I I input rates ( m g . k g 1. d a y - l) (r = + 0.49,p < 0.05, r = + 0.67,p < 0.01) in Type 1 diabetic men but non-diabetic in normal m e n (r = 0.24, r = 0.29, NS).
2(
1001"%
~~ 5
"-
.
100
Lipolytic enzymes, HD L and apo A metabolism 2~
10
2
4
6
8
10
12
lP4
Days
Fig.2. Radioactive decay curves of ~25I-labelled apolipoprotein (apo) A-I and mI-labelled apo A-II in three individuals with either high (apo A-I = 156 mg/dl and apo A-II = 29 mg/dl, upper panel), moderate (apo A-I = 137 mg/dl and apo A-II = 31 mg/dl, middle panel) or low (apo A-I = 109 mg/dl and apo A-II = 24 mg/dl, lower panel) plasma concentration of apo A-I
A-I + A - I I ratio). Mean values for apo A input rate for the Type 1 diabetic patients and control subjects are given in Table 3. The input of apo A - I and L p A - I particles (U9, Table 3, Figs. 1 and 4) was similar in the two groups. In contrast, the input rates for apo A - I (U1, Figs. 1 and 4) and apo A - I I (Us, Figs. 1 and 4) into L p A - I / A - I I were reduced in Type 1 diabetic patients.
Interrelations between apo A metabolism and plasma H D L and V L D L levels H D L cholesterol correlated inversely with FCRs of apo A-I, apo A - I I (r = - 0 . 9 1 in Type i diabetic patients, r = - 0 . 8 3 in control subjects; p < 0.001 in both groups) and the calculated F C R for L p A - I (r = - 0.08 in Type 1 diabetic patients, r = - 0.72 in control subjects, p < 0.001 in both groups). These correlations were accounted for by the variation of HDL2 but not HDL3. T h e r e was no association between H D L cholesterol levels and any of the input rates for apo A - I or apo A - I I in non-diabetic or Type i diabetic men (Tables 2 and 3). The concentration of apo A - I was determined by the rate of catabolism of the protein (plasma apo A - I vs apo A - I FCR, r = -0.82, p
